33b457a1c91e57d4de1193ccbdd427dd

This model is a fine-tuned version of facebook/opt-6.7b on the nyu-mll/glue [sst2] dataset. It achieves the following results on the evaluation set:

• Loss: 0.7750
• Data Size: 1.0
• Epoch Runtime: 817.6334
• Accuracy: 0.7859
• F1 Macro: 0.7857
• Rouge1: 0.7847
• Rouge2: 0.0
• Rougel: 0.7859
• Rougelsum: 0.7859

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro	Rouge1	Rouge2	Rougel	Rougelsum
No log	0	0	1.1169	0	2.7925	0.4884	0.3379	0.4884	0.0	0.4884	0.4884
No log	1	2104	0.8378	0.0078	10.0779	0.5093	0.3374	0.5093	0.0	0.5093	0.5081
No log	2	4208	0.7417	0.0156	27.8648	0.5266	0.4047	0.5266	0.0	0.5266	0.5266
0.0196	3	6312	0.7644	0.0312	56.2378	0.5799	0.5023	0.5799	0.0	0.5799	0.5787
0.4924	4	8416	0.4683	0.0625	88.4913	0.7928	0.7927	0.7917	0.0	0.7928	0.7940
0.4069	5	10520	0.5063	0.125	121.6539	0.7836	0.7799	0.7830	0.0	0.7847	0.7824
0.3124	6	12624	0.4481	0.25	216.3907	0.8171	0.8171	0.8160	0.0	0.8160	0.8177
0.2551	7	14728	0.4696	0.5	421.9573	0.8044	0.8039	0.8044	0.0	0.8044	0.8044
0.2074	8.0	16832	0.5031	1.0	822.0675	0.7917	0.7906	0.7917	0.0	0.7917	0.7917
0.1439	9.0	18936	0.6701	1.0	816.7059	0.7836	0.7830	0.7818	0.0	0.7836	0.7836
0.1038	10.0	21040	0.7750	1.0	817.6334	0.7859	0.7857	0.7847	0.0	0.7859	0.7859

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1